Power station screen cabinet equipment temperature early warning method and device and computer equipment

Abstract

The invention relates to a power station screen cabinet equipment temperature early warning method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a dual-light image of power station screen cabinet equipment to be analyzed according to acquisition parameters corresponding to current acquisition; the dual-light image comprises an infrared thermal imaging image and a visible light image; performing preprocessing, feature extraction processing, spatial registration processing and information fusion processing on the dual-light image to generate a fused image; performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information; and according to the temperature field distribution information and the hot spot area information, performing temperature analysis on the power station screen cabinet equipment to obtain temperature early warning information of the power station screen cabinet equipment. By adopting the method, the temperature early warning accuracy of the power station screen cabinet equipment can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and device for temperature early warning of power station cabinet equipment, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Power station cabinet equipment is a key equipment in the power system, mainly used for the control, protection and monitoring of the power system. These devices generate a lot of heat during operation. If the temperature is too high, it may cause equipment failure, insulation aging and even fire accidents. Therefore, real-time temperature monitoring and early warning of power station cabinet equipment has important practical significance and is an important means to ensure the safe and stable operation of the power system.

[0003] Traditional temperature monitoring methods mostly rely on a single temperature sensor or manual inspection. This method has problems such as untimely response, insufficient detection accuracy, and difficulty in comprehensive coverage. Especially during the operation of power station cabinet equipment, local overheating or failure may occur. Traditional monitoring methods often cannot effectively identify potential temperature anomalies, resulting in poor accuracy of temperature warning for power station cabinet equipment. Summary of the invention

[0004] Based on this, it is necessary to provide a temperature warning method, device, computer equipment, computer-readable storage medium and computer program product for power station cabinet equipment, which can improve the accuracy of temperature warning for power station cabinet equipment in response to the above technical problems.

[0005] In a first aspect, the present application provides a temperature early warning method for power station cabinet equipment, comprising:

[0006] According to the acquisition parameters corresponding to this acquisition, a dual-light image of the power station cabinet equipment to be analyzed is acquired; the dual-light image includes an infrared thermal imaging image and a visible light image;

[0007] Performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image;

[0008] Performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information;

[0009] According to the temperature field distribution information and the hot spot area information, the temperature of the power station cabinet equipment is analyzed to obtain temperature warning information of the power station cabinet equipment.

[0010] In one embodiment, before acquiring the bi-optical image of the power station cabinet device to be analyzed according to the acquisition parameters corresponding to the current acquisition, the method further includes:

[0011] In the case where the current acquisition is the first acquisition, the acquisition parameters corresponding to the lowest level of temperature anomaly are used as the acquisition parameters of the current acquisition;

[0012] In the case that the current acquisition is not the first acquisition, the acquisition parameters corresponding to the temperature anomaly level determined in the previous acquisition are used as the acquisition parameters of the current acquisition.

[0013] In one embodiment, the preprocessing, feature extraction, spatial registration and information fusion of the bi-optical image to generate a fused image includes:

[0014] Preprocessing the bi-optical image to obtain a preprocessed bi-optical image; the preprocessing at least includes quality assessment processing, image enhancement processing and region segmentation processing;

[0015] Extracting multi-scale local features and regional features from the preprocessed bi-optical image to construct feature description information;

[0016] According to the feature description information, registering the preprocessed bi-optical image to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image;

[0017] The pre-processed bi-optical image is fused according to the affine transformation matrix to generate a fused image.

[0018] In one embodiment, the pre-processed bi-optical image is registered according to the feature description information to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image, including:

[0019] According to the region segmentation result of the bi-optical image, initial registration points are screened out from the feature points corresponding to the feature description information; the region segmentation result is obtained by the region segmentation processing;

[0020] Filtering candidate registration points from the initial registration points according to the feature description information;

[0021] A random sampling consistency model is used to eliminate erroneous registration points from the candidate registration points to obtain target registration points;

[0022] The affine transformation matrix is ​​calculated and determined according to the target registration point.

[0023] In one embodiment, the step of fusing the pre-processed bi-optical image according to the affine transformation matrix to generate a fused image includes:

[0024] Obtaining a registered infrared thermal imaging image according to the affine transformation matrix and the infrared thermal imaging image;

[0025] Acquire an edge map of the visible light image, and acquire a temperature gradient map of the registered infrared thermal imaging image;

[0026] Determining a fusion weight of the bi-optical image according to the edge map and the temperature gradient map;

[0027] According to the fusion weight, the registered infrared thermal imaging image and the visible light image are fused to obtain a fused image.

[0028] In one embodiment, the temperature field reconstruction and hot spot detection processing are performed on the fused image to obtain temperature field distribution information and hot spot area information, including:

[0029] Obtaining initial temperature field distribution information according to the registered infrared thermal imaging image;

[0030] The temperature field distribution information is obtained according to the initial temperature field distribution information, the temperature gradient map, the fused image, the preset reference temperature and the preset power station panel cabinet equipment area importance weight;

[0031] According to the temperature field distribution information and the regional segmentation result of the bi-optical image, the hot spot region information is determined by adopting a threshold judgment method; the regional segmentation result is obtained through the regional segmentation process.

[0032] In one embodiment, the step of performing temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature warning information of the power station cabinet equipment includes:

[0033] Obtain the historical temperature field distribution information and historical hot spot area information corresponding to the historical collection;

[0034] Determine temperature magnitude change trend prediction information and hot spot area expansion trend information based on the historical temperature field distribution information, the historical hot spot area information, the temperature field distribution information and the hot spot area information;

[0035] Determine the temperature anomaly level collected this time according to the temperature magnitude change trend prediction information, the hot spot area expansion trend information and the temperature field distribution information;

[0036] Temperature warning information for the power station panel cabinet equipment is generated based on the temperature anomaly level collected this time, the temperature change trend prediction information, the hot spot area expansion trend information and the temperature field distribution information.

[0037] In a second aspect, the present application also provides a temperature early warning device for a power station cabinet, comprising:

[0038] An image acquisition module, used to acquire a bi-optical image of the power station cabinet equipment to be analyzed according to acquisition parameters corresponding to this acquisition; the bi-optical image includes an infrared thermal imaging image and a visible light image;

[0039] An image fusion module, used for performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image;

[0040] A temperature detection module is used to perform temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information;

[0041] The temperature analysis module is used to perform temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature warning information of the power station cabinet equipment.

[0042] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0043] According to the acquisition parameters corresponding to this acquisition, a dual-light image of the power station cabinet equipment to be analyzed is acquired; the dual-light image includes an infrared thermal imaging image and a visible light image;

[0044] Performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image;

[0045] Performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information;

[0046] According to the temperature field distribution information and the hot spot area information, the temperature of the power station cabinet equipment is analyzed to obtain temperature warning information of the power station cabinet equipment.

[0047] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0048] According to the acquisition parameters corresponding to this acquisition, a dual-light image of the power station cabinet equipment to be analyzed is acquired; the dual-light image includes an infrared thermal imaging image and a visible light image;

[0049] Performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image;

[0050] Performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information;

[0051] According to the temperature field distribution information and the hot spot area information, the temperature of the power station cabinet equipment is analyzed to obtain temperature warning information of the power station cabinet equipment.

[0052] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0053] According to the acquisition parameters corresponding to this acquisition, a dual-light image of the power station cabinet equipment to be analyzed is acquired; the dual-light image includes an infrared thermal imaging image and a visible light image;

[0054] Performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image;

[0055] Performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information;

[0056] According to the temperature field distribution information and the hot spot area information, the temperature of the power station cabinet equipment is analyzed to obtain temperature warning information of the power station cabinet equipment.

[0057] The above-mentioned power station cabinet equipment temperature warning method, device, computer equipment, computer-readable storage medium and computer program product first collect and obtain the dual-light image of the power station cabinet equipment to be analyzed according to the collection parameters corresponding to this collection, wherein the dual-light image includes an infrared thermal imaging image and a visible light image, and a comprehensive perception of the equipment status is achieved by combining the two types of image information. The infrared image provides the temperature distribution on the surface of the equipment to help detect temperature anomalies, and the visible light image provides the appearance structure and detailed information of the equipment, which is helpful for locating equipment faults, and can overcome the problem of limited monitoring range of a single sensor to ensure full coverage monitoring of the equipment status. In addition, dynamic adjustment of collection parameters can optimize resource utilization and ensure efficient collection in key areas; then, the dual-light image is preprocessed, feature extracted, spatially registered and information fused to generate a fused image, and the collected dual-light image is preprocessed to improve image quality, reduce noise and enhance contrast, and multi-scale local features and regional features are obtained through feature extraction to ensure that key information in the image is complete. The infrared image and the visible light image are retained and the spatial registration technology is used to accurately align the infrared image and the visible light image to ensure the consistency of the two data. Finally, the two image information is comprehensively processed through information fusion to generate a fused image, which enhances the reliability and accuracy of the image data and lays a solid foundation for the subsequent temperature field reconstruction and hotspot detection. Then, the fused image is processed for temperature field reconstruction and hotspot detection to obtain the temperature field distribution information and hotspot area information, which improves the system's ability to detect temperature anomalies, especially the ability to identify potential risks such as local overheating, and provides key support for temperature early warning. Finally, according to the temperature field distribution information and hotspot area information, the temperature of the power station cabinet equipment is analyzed to obtain the temperature early warning information of the power station cabinet equipment. Through the overall distribution of the temperature field and the characteristics of the hotspot area (such as temperature value, position, size, etc.), the temperature anomaly level of the equipment is accurately evaluated, and temperature early warning information is generated. The early warning information includes the possible abnormal areas of the equipment and the corresponding temperature status, which helps maintenance personnel take timely measures, thereby effectively reducing the risk of equipment failure. In the above method, the spatial registration technology of infrared thermal imaging images and visible light images is used to achieve precise alignment of temperature data and equipment structure information; through image preprocessing and information fusion, high-quality fused images are generated, which provides a reliable basis for temperature field reconstruction. Hotspot detection uses threshold judgment and region segmentation to accurately locate high-temperature areas and quickly identify potential anomalies. Finally, based on temperature distribution and hotspot information, the temperature anomaly level is accurately assessed and warning information is generated, realizing comprehensive monitoring of the equipment temperature status and precise warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0059] FIG1 is a schematic flow chart of a temperature early warning method for power station cabinet equipment in one embodiment;

[0060] FIG2 is a schematic diagram of a flow chart of a fused image generation step in one embodiment;

[0061] FIG3 is a schematic diagram of a temperature early warning system for power station cabinet equipment in one embodiment;

[0062] FIG4 is a structural block diagram of a temperature early warning device for power station cabinet equipment in one embodiment;

[0063] FIG. 5 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] In one embodiment, as shown in FIG1 , a temperature warning method for power station cabinet equipment is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, and tablet computers. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. In this embodiment, the method includes the following steps:

[0066] Step S101, acquiring a bi-optical image of a power station panel cabinet device to be analyzed according to acquisition parameters corresponding to this acquisition.

[0067] Among them, the dual-light image includes infrared thermal imaging image and visible light image.

[0068] Among them, the acquisition parameters refer to the camera parameters set by the terminal during the acquisition process, including the temperature range and emissivity of the infrared camera and the exposure time and white balance parameters of the visible light camera. These parameters are adjusted and configured according to the operating environment and monitoring requirements of the power station cabinet equipment.

[0069] Exemplarily, the terminal uses the built-in dual-light camera module to simultaneously start the infrared thermal imaging camera and the visible light camera, and images the target device according to the set acquisition parameters. The terminal adjusts the acquisition parameters according to the real-time operating status of the device, for example, dynamically adjusts the emissivity of the infrared thermal imaging camera according to the maximum temperature range of the device surface to ensure the accuracy of temperature measurement. At the same time, the terminal adjusts the exposure time and white balance parameters of the visible light camera in real time according to the ambient light conditions to obtain a clear visible light image. Finally, the terminal completes the synchronous acquisition of the dual-light image and transmits it to the subsequent processing module.

[0070] Step S102 , performing preprocessing, feature extraction, spatial registration and information fusion processing on the bi-optical image to generate a fused image.

[0071] Among them, preprocessing refers to the operations such as noise removal, contrast enhancement, and brightness adjustment on the collected bi-optical images to improve the image quality and provide clear and stable input data for subsequent analysis. Feature extraction processing is to extract key local features and regional features from the image, including texture, shape, edge and other information. These features help to accurately identify the various components of the equipment and potential problem areas. Spatial registration processing refers to aligning the infrared thermal imaging image and the visible light image so that the temperature information and structural information of the two can be accurately matched, thereby avoiding deviations between the images. Information fusion processing is to fuse the registered images, combine the temperature data of the infrared image with the structural information of the visible light image, and generate the final fused image, which provides an accurate data basis for subsequent temperature analysis and hotspot detection.

[0072] Exemplarily, the terminal first preprocesses the collected bi-optical image, removes noise using Gaussian filtering or median filtering, and enhances the image contrast through histogram equalization. Then, an image processing algorithm, such as Scale-Invariant Feature Transform (SIFT) or Speeded-Up Robust Features (SURF) algorithm, is used to extract important feature points and region descriptors from the preprocessed image, which will serve as the basis for subsequent registration and matching. In the spatial registration stage, the terminal accurately aligns the infrared thermal imaging image and the visible light image through a feature point matching algorithm, such as the Random Sample Consensus (RANSAC) algorithm, to ensure that the spatial correspondence between the two is accurate. Finally, the terminal merges the registered images through a weighted fusion technique, and generates a fused image using the temperature information in the infrared image and the structural information in the visible light image, which has higher temperature detection accuracy and structural details.

[0073] Step S103, performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information.

[0074] Among them, temperature field reconstruction refers to reconstructing the temperature field on the surface of the equipment through interpolation algorithms and heat conduction models based on the temperature data in the fused image. The purpose of temperature field reconstruction is to expand discrete temperature data points into a complete temperature distribution map, which can provide temperature change trends in various areas on the surface of the equipment. Hotspot detection is to identify areas with abnormal temperatures, that is, hotspot areas where the temperature exceeds the preset safety range, by setting temperature thresholds or using regional statistical analysis. Hotspot area information includes the spatial location, area, temperature value and other characteristics of the hotspot, which are crucial for equipment maintenance and fault warning.

[0075] Exemplarily, the terminal interpolates the temperature information of the fused image (such as bilinear interpolation or kriging interpolation) to smooth the temperature data points in the image and generate a continuous temperature field distribution map. In addition, the terminal uses temperature gradient analysis (TemperatureGradient Analysis) and local statistical analysis (Local Statistical Analysis) to identify areas with excessively high temperatures in the device, i.e., hot spots, by setting temperature thresholds or standard deviation-based anomaly detection methods (such as the Z-score method). Through these algorithms, the terminal can accurately identify and mark potential fault areas in the device, providing key data for subsequent early warning systems.

[0076] Step S104, performing temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information, and obtaining temperature warning information of the power station cabinet equipment.

[0077] Among them, temperature analysis refers to the evaluation of the overall temperature status of power station cabinet equipment based on temperature field distribution information and hot spot area information. By analyzing the temperature values ​​of various areas on the surface of the equipment and the expansion trend of hot spots, it is possible to accurately determine whether the equipment has potential temperature anomalies or failures. Temperature warning information is generated based on the analysis results, including the severity of the current temperature anomaly, the location of the abnormal area, the temperature value, and the predicted temperature change trend, providing timely and accurate warning data for operation and maintenance personnel.

[0078] Exemplarily, the terminal first evaluates the overall temperature status of the device and determines whether there is overheating on the surface of the device by performing global and local analysis of the temperature field distribution information. Combined with the hot spot area information, the terminal identifies the local overheating area and calculates whether its temperature exceeds the preset safety threshold (such as the maximum safe temperature of the device). In addition, based on the changing trend of the device temperature (by comparing the currently collected temperature data with historical data) and the expansion of the hot spot area, the terminal predicts whether the device temperature may continue to rise or fail, and generates temperature warning information. The warning information includes the temperature anomaly level (such as mild, moderate, and severe anomalies), the spatial location and temperature data of the abnormal area, and provides operational suggestions (such as whether shutdown, cooling, or maintenance is required).

[0079] In the above-mentioned power station cabinet equipment temperature warning method, first, according to the acquisition parameters corresponding to this acquisition, the dual-light image of the power station cabinet equipment to be analyzed is acquired, wherein the dual-light image includes an infrared thermal imaging image and a visible light image, and a comprehensive perception of the equipment status is achieved by combining the two types of image information. The infrared image provides the temperature distribution on the surface of the equipment to help detect temperature anomalies, and the visible light image provides the appearance structure and detailed information of the equipment, which is helpful for locating equipment faults and can overcome the problem of limited monitoring range of a single sensor to ensure full coverage monitoring of the equipment status. In addition, dynamic adjustment of acquisition parameters can optimize resource utilization and ensure efficient acquisition in key areas; then, the dual-light image is preprocessed, feature extracted, spatially registered and information fused to generate a fused image, and the acquired dual-light image is preprocessed to improve image quality, reduce noise and enhance contrast, and multi-scale local features and regional features are obtained through feature extraction to ensure that key information in the image is fully retained, and the infrared image is processed using spatial registration technology. The external image and the visible light image are precisely aligned to ensure the consistency of the two data. Finally, through information fusion, the two image information is comprehensively processed to generate a fused image, which enhances the reliability and accuracy of the image data and lays a solid foundation for subsequent temperature field reconstruction and hotspot detection. Then, the fused image is processed for temperature field reconstruction and hotspot detection to obtain temperature field distribution information and hotspot area information, which improves the system's ability to detect temperature anomalies, especially the ability to identify potential risks such as local overheating, and provides key support for temperature warning. Finally, based on the temperature field distribution information and hotspot area information, the power station panel cabinet equipment is temperature analyzed to obtain the temperature warning information of the power station panel cabinet equipment. Through the overall distribution of the temperature field and the characteristics of the hotspot area (such as temperature value, position, size, etc.), the temperature anomaly level of the equipment is accurately evaluated, and temperature warning information is generated. The warning information includes possible abnormal areas of the equipment and the corresponding temperature status, which helps maintenance personnel take timely measures, thereby effectively reducing the risk of equipment failure. In the above method, the spatial registration technology of infrared thermal imaging images and visible light images is used to achieve precise alignment of temperature data and equipment structure information; through image preprocessing and information fusion, high-quality fused images are generated, which provides a reliable basis for temperature field reconstruction. Hotspot detection uses threshold judgment and region segmentation to accurately locate high-temperature areas and quickly identify potential anomalies. Finally, based on temperature distribution and hotspot information, the temperature anomaly level is accurately assessed and warning information is generated, realizing comprehensive monitoring of the equipment temperature status and precise warning.

[0080] In an exemplary embodiment, before acquiring the dual - light image of the power station cabinet equipment to be analyzed according to the acquisition parameters corresponding to this acquisition in the above - mentioned step S101, it further includes: in the case where this acquisition is the first acquisition, taking the acquisition parameters corresponding to the lowest - level temperature anomaly level as the acquisition parameters for this acquisition; in the case where this acquisition is not the first acquisition, taking the acquisition parameters corresponding to the temperature anomaly level determined in the previous acquisition as the acquisition parameters for this acquisition.

[0081] Among them, the temperature anomaly level is an index classified according to the severity of the equipment temperature state. The acquisition parameters include the setting parameters of the infrared camera and the visible - light camera, such as the temperature range, emissivity, exposure time, white balance, etc. of the infrared image. The setting of these parameters directly affects the quality of the acquired image and the accuracy of the temperature data.

[0082] Exemplarily, the terminal first determines whether this acquisition is the first acquisition. If it is the first acquisition, the temperature anomaly level is set to the lowest level (i.e., the "normal" or slightly abnormal level), and the corresponding acquisition parameters are set according to this level, such as a lower temperature range of the infrared thermal imaging camera and standard exposure settings, to ensure that the acquisition parameters are not too sensitive. For subsequent acquisitions, the terminal will adjust the parameters of this acquisition according to the temperature anomaly level determined in the previous acquisition. If the previous acquisition result shows that the temperature anomaly is moderate or severe, the terminal will adjust the acquisition parameters, such as increasing the temperature range, increasing the exposure time, etc., in order to capture more detailed temperature information, so as to provide higher - precision data support for subsequent temperature analysis and anomaly detection.

[0083] Specifically, the temperature anomaly level can be simply judged by the highest temperature, that is, which temperature range of the temperature anomaly level the highest temperature reaches, it is that temperature anomaly level. When the temperature T < T1, it is in the normal temperature range, and the reference frequency f0 is adopted; when T1 ≤ T < T2, it is a mild anomaly, and 2 times the reference frequency is adopted; when T2 ≤ T < T3, it is a moderate anomaly, and 4 times the reference frequency is adopted; when T ≥ T3, it is a severe anomaly, and 8 times the reference frequency is adopted; where T1, T2, and T3 are preset temperature thresholds; for example, set T1 = 45 °C, T2 = 65 °C, T3 = 85 °C, and f0 = 5 min / time, then the acquisition frequency is adjusted as follows: in the normal state, f = f0 = 5 minutes / time; in the mild - anomaly state, f = 2f0 = 2.5 minutes / time; in the moderate - anomaly state, f = 4f0 = 1.25 minutes / time; in the severe - anomaly state, f = 8f0 = 0.625 minutes / time.

[0084] The parameter adjustment of the dual - light camera also determines the specific parameter values according to the real - time feedback. Among them, the adjustable parameters of the infrared camera include the temperature measurement range and emissivity; the adjustable parameters of the visible - light camera include the exposure time and ISO sensitivity.

[0085] Furthermore, the image may be collected by collecting 3-5 groups of dual-light images, each group of data including an infrared thermal imaging image, a visible light image, and a collection timestamp.

[0086] In this embodiment, by dynamically adjusting the collection parameters, the collection process can be intelligently optimized according to the real-time temperature status and abnormality level of the device to ensure the accuracy and reliability of the temperature data, avoid excessive data collection when the device temperature is normal, reduce unnecessary waste of resources, and at the same time, improve the data sensitivity and detection accuracy when the device temperature is abnormal.

[0087] In an exemplary embodiment, as shown in FIG. 2 , the above step S102 performs preprocessing, feature extraction, spatial registration and information fusion processing on the bi-optical image to generate a fused image, which can also be achieved by the following steps:

[0088] Step S201 , preprocessing the bi-optical image to obtain a pre-processed bi-optical image.

[0089] Step S202, extracting multi-scale local features and regional features from the pre-processed bi-optical image to construct feature description information.

[0090] Step S203 , performing registration processing on the pre-processed bi-optical image according to the feature description information, and obtaining an affine transformation matrix of the infrared thermal imaging image relative to the visible light image.

[0091] Step S204: performing fusion processing on the pre-processed bi-optical image according to the affine transformation matrix to generate a fused image.

[0092] The preprocessing includes at least quality assessment processing, image enhancement processing and region segmentation processing.

[0093] Among them, dual-light image preprocessing refers to the necessary image processing of the collected infrared thermal imaging images and visible light images to improve image quality and remove possible noise. Quality assessment is the process of evaluating image quality to ensure that the temperature information in the image is clear and reliable. Indicators such as signal-to-noise ratio and contrast are often used for evaluation. Image enhancement includes contrast enhancement, brightness adjustment, edge sharpening, etc., which aims to highlight important information in the image and improve the effect of feature extraction. Regional segmentation methods (such as superpixel segmentation, edge-based segmentation algorithms, etc.) divide the image into multiple regions, which facilitates independent analysis of temperature characteristics of different regions during subsequent processing.

[0094] For example, the terminal evaluates the quality of the collected dual-light images, uses algorithms such as peak signal-to-noise ratio and structural similarity to evaluate the quality of the images, and removes noise in the images through Gaussian filtering or median filtering. Then, histogram equalization is used to enhance the contrast of the image, making the temperature information more prominent, and superpixel segmentation is used to divide the image into regions, thereby extracting device regions with different functions.

[0095] In the feature extraction stage, the terminal extracts multi-scale local features and regional features from the preprocessed image. Common feature extraction methods include SIFT and SURF, as well as regional features such as local binary patterns. These feature description information is used for subsequent image registration and fusion processing.

[0096] In the spatial registration processing stage, the terminal uses a feature point matching algorithm (such as SIFT or SURF) to match the infrared thermal imaging image and the visible light image and calculate the affine transformation matrix between them. By applying the RANSAC algorithm, the matching results are optimized to obtain accurate registration results.

[0097] Finally, the terminal fuses the infrared thermal imaging image with the visible light image according to the calculated affine transformation matrix. Information fusion methods include weighted averaging and pixel-level fusion, ensuring that the final fused image can retain both temperature data (infrared image) and structural data (visible light image), providing high-quality data support for subsequent temperature field reconstruction and hotspot detection.

[0098] Specifically, ① calculate the quality score of the bi-optical image: .

[0099] Among them, SNR(I) is the signal-to-noise ratio, E(I) is the edge clarity, C(I) is the contrast, w1, w2, w3 are weight coefficients and w1+w2+w3=1. For example, w1=0.4, w2=0.3, w3=0.3, the signal-to-noise ratio SNR is calculated using the signal-noise separation method based on Gaussian filtering. The edge clarity is based on the image gradient amplitude statistics. The contrast uses the Michelson contrast of the local area.

[0100] Determine whether the quality score Q(I) of the bi-optical image is less than the threshold Q thre If yes, then perform image enhancement and then perform region segmentation. If no, then perform region segmentation directly. For example, Q thre Can be set to 0.6.

[0101] ② Perform regional segmentation on the bi-optical image and extract the outline of the cabinet equipment based on the level set method; use the multi-scale superpixel segmentation algorithm to divide the image into several sub-regions; combine prior knowledge to identify key functional areas and construct a regional importance weight map W(x, y) =γ1·W temp (x,y)+γ2·W str (x,y), where W temp is the temperature importance weight, W str is the structural importance weight, γ1 and γ2 are weight coefficients, and the sum is 1. Combined with the structural design and working principle of the power station equipment, determine which areas are critical to the reliability and stability of the equipment, such as power modules, control panels, and heat dissipation areas. They are critical to the normal operation of the equipment, so the weights of these areas are higher. Among them, the energy function of the level set method is designed as: E(φ)=μ·E reg (φ)+ν·E area (φ)+λ·E edge (φ), where E reg is the regularization term, E area is the area term, E edge is the edge term, μ, ν, λ are weight coefficients.

[0102] A physical constraint model is established based on the structural characteristics of the power station cabinet equipment, including the relative position relationship of equipment components and the spatial distribution of key functional areas; the position relationship of each component of the equipment in the image provides a spatial constraint for temperature field reconstruction. By modeling the positions of these components, the temperature distribution can be optimized based on the known equipment structure. For example, the power module is usually located at the bottom of the equipment or near the heat dissipation area, while the control board and sensor are usually located at the top or in the middle of the equipment.

[0103] ③The enhancement process of visible light images includes histogram equalization and contrast enhancement; the enhancement process of infrared thermal imaging images includes temperature normalization and denoising.

[0104] Histogram equalization includes: setting the contrast limit threshold α=3.0, the histogram exceeding the threshold is cropped and redistributed; the image is divided into 8×8 sub-blocks, each sub-block is histogram equalized, and bilinear interpolation is used to merge the block boundaries. Contrast enhancement includes: using piecewise linear transformation, grayscale mapping: g(x)=α·f(x)+β, where α is the contrast gain, ranging from [1.2-1.8], β is the brightness adjustment parameter, ranging from [-30, 30], and the parameter is adaptively adjusted according to the statistical characteristics of the image. Temperature normalization is: In(x, y)=(I(x, y)-T min ) / (T max -T min ), T min is the minimum temperature value of the current image, Tmax is the maximum temperature value of the current image. The denoising process is performed using a median filter.

[0105] ④ Extract multi-scale local features of bi-optical images, including: extracting SIFT feature points and their 128-dimensional descriptors for visible light images; extracting SURF feature points and their 64-dimensional descriptors for infrared images; extracting Harris corner point features for bi-optical images;

[0106] Extracting regional features of bi-optical images, including: extracting shape descriptors based on the HOG operator; extracting temperature distribution pattern features of infrared images, including temperature gradient direction histograms and hot spot distribution features; extracting texture features based on the LBP operator; specifically, hot spot distribution features are segmented by thresholds, and connected regions are marked, and a spatial distribution description is formed through centroid calculation, including hot spot density maps, hot spot distance statistics, and shape features.

[0107] Construct context-aware feature descriptor D: .

[0108] in, For local feature description, SIFT / SURF descriptors and Harris corner features are integrated; Neighborhood structure information, including HOG shape descriptor, temperature distribution pattern features and LBP texture features, is used to describe the structural characteristics of the area around the feature point; It is a constraint item based on prior knowledge, which contains regional importance weight and physical constraint information.

[0109] Feature descriptor D achieves an organic combination of multi-scale and multi-modal features by fusing local feature descriptions (SIFT / SURF and Harris features), contextual structure information (HOG shape descriptor, temperature distribution features and LBP texture features) and constraints based on prior knowledge (regional importance weights and physical constraint information). It not only ensures the accuracy of feature matching, but also makes full use of regional structure information and domain expertise, thereby improving the accuracy and reliability of image registration and anomaly detection of power station cabinet equipment.

[0110] In this embodiment, through the preprocessing, feature extraction, spatial registration and information fusion processing of the dual-light image, the image data can be optimized and accurately aligned, and the errors and noise that may be generated during the image acquisition process can be eliminated. By effectively fusing the temperature data of the infrared image with the structural information of the visible light image, the fused image finally generated provides accurate and comprehensive data support for the subsequent temperature field reconstruction and hotspot detection, significantly improving the precision and accuracy of temperature monitoring. Ultimately, the real-time temperature monitoring capability of the power station cabinet equipment is improved, especially in identifying local temperature anomalies and potential fault areas with high reliability.

[0111] In an exemplary embodiment, the above-mentioned step S203 performs registration processing on the pre-processed bi-optical image according to the feature description information to obtain the affine transformation matrix of the infrared thermal imaging image relative to the visible light image, and also includes: according to the regional segmentation result of the bi-optical image, initial registration points are screened out from the feature points corresponding to the feature description information; the regional segmentation result is obtained by regional segmentation; according to the feature description information, candidate registration points are screened out from the initial registration points; a random sampling consistency model is used to eliminate erroneous registration points from the candidate registration points to obtain target registration points; and according to the target registration points, the affine transformation matrix is ​​calculated and determined.

[0112] Exemplarily, the terminal first divides the infrared thermal imaging image and the visible light image into different regions by regional segmentation of the dual light image. In these regions, key feature points (such as corner points, edge points, etc.) are extracted, and initial registration points are screened out according to the feature descriptors of each region. These initial registration points serve as the starting point of registration, and candidate registration points with a high degree of matching are further screened out through the feature matching algorithm. Then, the terminal uses the RANSAC algorithm for optimization to eliminate incorrect matching points. RANSAC randomly selects a subset from the candidate points, calculates a preliminary affine transformation matrix, and screens the optimal registration point by verifying the applicability of the transformation matrix to other points. Finally, the target registration point is selected, and an accurate affine transformation matrix is ​​calculated. The affine transformation matrix contains transformation parameters such as translation, rotation, and scaling, which can accurately align the infrared thermal imaging image to the visible light image. Through this matrix, the terminal can accurately align the infrared thermal imaging image and the visible light image, ensuring that subsequent image analysis is based on consistent spatial coordinates, thereby providing a reliable data basis for temperature field reconstruction and hotspot detection.

[0113] Specifically, the visible light image can provide clear edge and structure information, so as the reference image for registration, the visible light image is selected as the reference image, and the infrared thermal imaging image is selected as the image to be registered.

[0114] ① Rough matching based on the regional segmentation results: Use the regional feature graph G(V, E) to divide the matching area, where the vertex set V represents the segmented area and the edge set E represents the spatial adjacency between regions. For example, regional connectivity analysis, through common edge determination, that is, whether the two regions share boundary pixels, the adjacency matrix is ​​constructed according to the common edge determination; the priority search area is determined according to the regional importance weight W(x, y); in the priority search area, the kd tree is used to perform the nearest neighbor search of the feature descriptor D to obtain the initial matching pair set M1; when searching for the nearest neighbor, k can take the value of 2, that is, take 2 nearest neighbors, the feature dimension of the kd tree is the dimension of D, and it is split according to the maximum variance dimension, and the leaf node capacity is a maximum of 10 feature points. When searching, the distance ratio threshold is set to 0.8, and the search radius is limited to 100 pixels.

[0115] ②Perform fine matching based on spatial constraints: calculate the Euclidean distance of feature descriptors as a similarity measure; construct spatial constraints SC: .

[0116] in, They represent two feature point matching pairs to be evaluated, each of which contains a feature point in the reference image and a feature point in the image to be registered. Represents the distance between feature point pairs, Represents the angle between the line connecting the feature point pairs and the horizontal direction, , is the weight coefficient; is the distance normalization, that is ; is the angle normalization, that is , , They are the maximum distance threshold and the maximum angle threshold respectively.

[0117] The initial matching pair set M1 is screened based on the spatial constraint SC to obtain the optimized matching pair set M2; for example, the SC threshold is set to 0.8, that is, matching pairs with SC less than 0.8 are retained.

[0118] ③ Use the RANSAC algorithm to estimate the affine transformation matrix H and define the confidence evaluation function as follows:

[0119]

[0120] in, is the cosine similarity of feature descriptor D; For point pair Conformity to the affine transformation matrix H; is a priori constraint based on the regional importance weight; , , is the weight coefficient and .

[0121] For example, the affine transformation matrix estimation process is: random sampling: select 3 pairs of matching points each time; calculate the affine matrix: 6 degrees of freedom; inlier statistics: calculate the projection error.

[0122] The final matching pair set M3 is obtained by screening according to the Score threshold, and the affine transformation matrix is ​​calculated using the final matching pair set M3 to complete the image registration. For example, the Score threshold can be set to 0.7, that is, the matching pairs with Score>0.7 are retained as the final matching pair set M3.

[0123] ④Then, the least squares method can be used to solve the parameters of the affine transformation matrix. Then, according to the affine transformation matrix, the pixels of the infrared thermal imaging image are transformed into the pixel coordinate system of the visible light image to complete the image registration.

[0124] For example, there are n pairs of matching points (xi, yi) and (x i ′,y i ′) (Each pair of matching points contains the coordinates of the visible light image and the coordinates of the infrared thermal imaging image). For each pair of matching points, the following two equations can be obtained:

[0125]

[0126] Among them, a, b, c, d, e, f are the affine transformation parameters to be determined.

[0127] Arrange these equations into matrix form: A·θ=B.

[0128]

[0129] Since the system of equations may be overdetermined (i.e. 2N>6), the least squares method is needed to solve the parameter vector θ. The formula for the least squares solution is: . T is the matrix transpose.

[0130] Alternatively, the problem can be solved by using methods such as QR decomposition (orthogonal triangular decomposition) and singular value decomposition (SVD).

[0131] Through the priority search mechanism guided by the regional feature map G(V, E) and the fine matching process based on the spatial constraint SC, combined with the regional importance weight W(x, y) and the physical characteristics of the equipment constraints, multiple constraints and optimizations of the registration process are realized; combined with the screening mechanism of the confidence evaluation function Score, the registration accuracy is guaranteed and the calculation complexity is reduced. It is particularly suitable for scenarios with specific structural characteristics such as power station cabinet equipment, and has strong robustness and practicality.

[0132] In this embodiment, high-precision image registration is achieved by combining region segmentation, feature description information screening and RANSAC optimization algorithm. Through the screening of initial registration points, the screening of candidate registration points and the optimization of RANSAC algorithm, the accurate alignment of infrared thermal imaging images and visible light images is ensured, and the registration errors caused by different image shooting angles or other factors are eliminated. The affine transformation matrix finally generated provides an accurate spatial coordinate basis for subsequent temperature field reconstruction and hotspot detection, thereby significantly improving the accuracy and reliability of temperature monitoring.

[0133] In an exemplary embodiment, the above-mentioned step S204 fuses the pre-processed bi-light image according to the affine transformation matrix to generate a fused image, and also includes: obtaining the registered infrared thermal imaging image according to the affine transformation matrix and the infrared thermal imaging image; obtaining the edge map of the visible light image, and obtaining the temperature gradient map of the registered infrared thermal imaging image; determining the fusion weight of the bi-light image according to the edge map and the temperature gradient map; and fusing the registered infrared thermal imaging image and the visible light image according to the fusion weight to obtain a fused image.

[0134] Exemplarily, the terminal first uses an affine transformation matrix to align the infrared thermal imaging image and accurately register it to the spatial coordinate system of the visible light image. Then, edge information is extracted by applying edge detection to the visible light image. The edge map is generated by extracting edge information from the visible light image, which can highlight the obvious structural boundaries and contours on the surface of the device. Edge information plays an important role in the fusion process and is usually used to determine which areas require high weights. The temperature gradient of the infrared image is calculated to generate a temperature gradient map. Then, the terminal calculates the weight of each image area through a weighted fusion algorithm. Hot spots and edge areas are assigned higher weights. Edge areas have higher structural importance, while areas with larger temperature gradients indicate temperature anomalies, ensuring that the key information of these areas is fully reflected in the final fused image. Finally, based on these fusion weights, the terminal fuses the infrared thermal imaging image with the visible light image to generate a high-quality fused image. Common fusion methods include weighted averaging, that is, adjusting the contribution ratio of the infrared image to the visible light image according to the importance (weight) of the area, so that the final fused image can retain both structural information (from the visible light image) and temperature information (from the infrared image). For example, the temperature information of the hot spot area may rely more on the infrared image, while the edge area may rely more on the visible light image. This image combines temperature information and structural information, providing an accurate data basis for subsequent temperature field reconstruction and hot spot detection.

[0135] Specifically, the registered infrared thermal imaging image IR and visible light image VIS are normalized to unify the image value range to the interval [0, 1], unify the numerical range of the two images, and eliminate the dimension difference.

[0136] Calculate the Sobel edge map Ed(x, y) of the visible light image. The Sobel operator calculates:

[0137] Horizontal gradient: ;

[0138] Vertical gradient: ;

[0139] Edge Strength: ,The edge map reflects the structural information and contour features in the visible light image.

[0140] Extract the temperature gradient map Tg(x, y) of the infrared thermal imaging image, horizontal temperature gradient: ; Vertical temperature gradient: ; Temperature gradient amplitude: ; The temperature gradient map reflects the severity of temperature changes in the infrared image.

[0141] Generate Adaptive Weights , μ1 and μ2 are adjustment coefficients and μ1+μ2=1; when setting, μ1<μ2 can be set, the high edge response area tends to retain visible light information, and the high temperature gradient area tends to retain infrared information.

[0142] Fusion image generation: .

[0143] Among them, F(x, y) is the fused image, IR(x, y) is the normalized infrared thermal imaging image, and VIS(x, y) is the normalized visible light image. Using this fusion formula, more visible light details are retained in the edge-significant area, and more infrared information is retained in the area with large temperature gradient, realizing the adaptive fusion of structural information and temperature information.

[0144] The dual-light image fusion algorithm organically combines the Sobel edge map Ed(x, y) and the temperature gradient map Tg(x, y) by introducing the adaptive weight Z(x, y), thereby realizing the complementary advantages of the structural details of the visible light image and the temperature information of the infrared image; through the adjustable weight coefficients μ1 and μ2, the system can flexibly adjust the fusion ratio of structural information and temperature information according to actual application requirements, which not only ensures the visual quality of the fused image, but also ensures the accurate transmission of temperature information.

[0145] In this embodiment, through the precise registration of the affine transformation matrix and the weighted fusion based on the edge map and the temperature gradient map, an accurate fusion image can be generated, which not only retains the temperature distribution information of the power station cabinet equipment, but also accurately presents the structural details of the equipment surface. The weighted fusion method ensures that the information of hot spots, edge areas and other key areas is retained first, improves the comprehensive information content and accuracy of the image, provides a reliable foundation for subsequent temperature field reconstruction, hot spot detection and temperature warning, and significantly improves the accuracy of equipment monitoring and fault prediction.

[0146] In an exemplary embodiment, the above step S103 performs temperature field reconstruction processing and hotspot detection processing on the fused image to obtain temperature field distribution information and hotspot area information, and also includes: obtaining initial temperature field distribution information based on the registered infrared thermal imaging image; obtaining temperature field distribution information based on the initial temperature field distribution information, temperature gradient map, fused image, preset reference temperature and preset power station cabinet equipment area importance weight; according to the temperature field distribution information and the regional segmentation result of the dual-light image, the hotspot area is determined by threshold judgment to obtain hotspot area information; the regional segmentation result is obtained by regional segmentation processing.

[0147] Exemplarily, the terminal first obtains preliminary temperature field distribution information through the registered infrared thermal imaging image, which shows the temperature changes of various areas on the surface of the device. The infrared thermal imaging image provides temperature information of various areas on the surface of the device. After preprocessing and registration, a preliminary temperature field distribution map can be obtained, which presents a rough distribution of the surface temperature of the device. Then, combined with the temperature gradient map and the fused image, the optimized temperature field distribution map is obtained by weighted calculation based on the preset reference temperature and regional importance weight. Considering the different functional areas of the device and the importance of each area, key areas (such as power supply, control board, etc.) are given higher weights to ensure that they receive more attention in the temperature field analysis. The preset reference temperature provides a standard reference temperature value for the device, which is used to compare the actual temperature of the device surface and identify areas with abnormal temperature. According to the temperature field distribution and the segmentation of the device area, the terminal uses the threshold judgment method to identify the area with too high temperature (i.e., the hot spot area), and outputs the hot spot area information including the hot spot location, temperature value and size.

[0148] Specifically, ① reconstruct the temperature field based on the fused image F(x,y), and the reconstruction function is: .

[0149] Wherein, T(x, y) is the reconstructed temperature field distribution function, which represents the actual temperature value at the point (x, y) in the plane coordinate system of the cabinet equipment; T base is the reference temperature value, which can be set to the ambient temperature; F(x, y) is the fused image; W(x, y) is the regional importance weight; T g (x, y) is the temperature gradient diagram; div represents the divergence operator; K(x, y) is the correction coefficient, , where K0 is the reference correction coefficient obtained according to camera calibration, G(x, y) is the normalized temperature gradient amplitude, L(x, y) is the local temperature uniformity, and β1 and β2 are weight coefficients.

[0150] Furthermore, the divergence operator acts on the temperature gradient graph T g (x, y), the calculation formula is: , where T g (x) and T g (y) are the temperature gradient components in the x and y directions respectively. This operator is used to describe the diffusion characteristics of the temperature field in space and reflects the degree of heat convergence or divergence.

[0151] Calculation of normalized temperature gradient amplitude G(x, y): First calculate the partial derivatives of temperature in the x and y directions: T x =∂T / ∂x,T y =∂T / ∂y, calculate the gradient amplitude: , normalized: , G(x, y) indicates the severity of temperature change. The larger the value, the faster the temperature changes.

[0152] Calculation of local temperature uniformity L(x, y): Calculate the local standard deviation within the local neighborhood N(x, y) of point (x, y): , where μ is the local mean; local uniformity: ; L(x, y) describes the smoothness of the temperature distribution. The larger the value, the more uniform the temperature distribution.

[0153] The reconstruction function realizes the mapping transformation from image grayscale value to actual temperature value by combining reference temperature, fused image information, regional weight and temperature gradient characteristics, while taking into account the spatial distribution characteristics of the temperature field.

[0154] ② Based on T(x, y), local statistical features are calculated, the threshold method is used to determine the hotspot area, and the characteristic parameters of the hotspot area, including the maximum temperature, area and shape characteristics, are extracted. A hotspot tracking table is established to record the position and temperature changes.

[0155] The criteria for determining hotspot areas are: .

[0156] Where H(x, y) represents the hotspot area; N(x, y) represents the local neighborhood; μ is the local mean; σ is the local standard deviation; κ is the detection coefficient; T thre is the preset temperature threshold.

[0157] Furthermore, when determining the local neighborhood, taking point (x, y) as the center, combined with the above regional segmentation results, considering the physical structure characteristics of the device, N(x, y) = {(i, j)|(i, j) belongs to the same superpixel area and |ix|≤r and |jy|≤r}, r is the search radius, for example, the value is 5-7 pixels, the same superpixel area refers to the result of the above multi-scale superpixel segmentation, requiring (i, j) and (x, y) to be in the same functional area. When calculating, determine the superpixel area where point (x, y) is located, take a (2r+1)×(2r+1) window in the area, calculate the temperature mean μ and standard deviation σ in the window, and judge whether T(x, y) exceeds the threshold.

[0158] In this embodiment, by combining temperature field reconstruction, temperature gradient map, region segmentation and hot spot detection methods, accurate monitoring of the temperature status of power station cabinet equipment can be achieved. The temperature field distribution information is optimized through comprehensive analysis of multiple factors, especially focusing on key areas and high-temperature areas, which significantly improves the ability to identify temperature anomalies. The hot spot detection method based on threshold judgment can quickly locate high-temperature areas in the equipment and provide early warning information in a timely manner, effectively avoiding equipment failures or catastrophic events.

[0159] In an exemplary embodiment, the above-mentioned step S104 performs temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature warning information of the power station cabinet equipment, and also includes: obtaining historical temperature field distribution information and historical hot spot area information corresponding to the historical collection; determining the temperature magnitude change trend prediction information and the hot spot area expansion trend information according to the historical temperature field distribution information, the historical hot spot area information, the temperature field distribution information and the hot spot area information; determining the temperature anomaly level of this collection according to the temperature magnitude change trend prediction information, the hot spot area expansion trend information and the temperature field distribution information; generating temperature warning information for the power station cabinet equipment according to the temperature anomaly level, the temperature magnitude change trend prediction information, the hot spot area expansion trend information and the temperature field distribution information collected this time.

[0160] Exemplarily, the terminal obtains the historical data of the historical temperature field distribution and the historical hot spot area, and analyzes the temperature change trend and the expansion trend of the hot spot area in combination with the current temperature field distribution information. The temperature change trend prediction information predicts whether the temperature will rise or fall in the future based on the comparison of the historical temperature data and the current temperature field distribution. The hot spot area expansion trend information predicts whether the hot spot area will further expand by comparing the temperature of the current hot spot area with the change trend of the historical hot spot area. Then, based on these data, the terminal will calculate the temperature anomaly level and generate corresponding early warning information. For example, if the current temperature change trend shows an upward trend and the hot spot area continues to expand, the terminal may issue a "severe anomaly" warning and recommend further cooling or equipment inspection measures. The temperature anomaly level is comprehensively evaluated based on the currently collected temperature field distribution and historical temperature field distribution data, combined with the temperature change trend prediction information and the hot spot area expansion trend information. By analyzing the speed of temperature change and the expansion of the hot spot area, the temperature anomaly level of this acquisition can be accurately judged, such as: normal, mild anomaly, moderate anomaly, severe anomaly, etc. Finally, based on the determined temperature anomaly level, temperature change trend, hotspot area expansion trend and temperature field distribution information, temperature warning information is generated. The temperature warning information will clearly indicate the abnormal degree of the current temperature of the equipment and provide targeted warning response suggestions, such as whether shutdown, cooling or other maintenance measures are required.

[0161] Specifically, the temperature field T(x, y) and hotspot detection results output by the temperature monitoring module are sorted in time series to establish a monitoring data sequence {T(x, y, t)}. The data is spatially weighted in combination with the regional importance weight W(x, y) to generate a regional temperature feature vector, including the maximum temperature, average temperature, temperature standard deviation, and hotspot area. Then, the sliding time window method is used to calculate the temperature change rate and acceleration, and exponential smoothing is used to predict the temperature trend, and a trend evaluation index is established: .

[0162] Where ΔT / Δt represents the rate of change of temperature over time, ∆ 2 T / Δt 2 represents the rate of change of the temperature change rate, P(t) is the temperature prediction deviation, ξ1, ξ2, ξ3 are weight coefficients and ξ1+ξ2+ξ3=1.

[0163] Further, , , the temperature prediction deviation P(t) is the temperature trend predicted by exponential smoothing, , T pre (t) is the exponential smoothing forecast value based on historical data, T act (t) is the actual measured value.

[0164] Next, based on multi-level temperature thresholds and trend evaluation metrics, an abnormal discrimination model is constructed. The discrimination basis includes the degree of temperature overrun, the duration of overrun, the change trend, and the spatial distribution characteristics, and the temperature anomaly level is output. The degree of temperature overrun is the comparison between the actual temperature and the preset thresholds T1, T2, and T3. The duration of overrun is the duration of the abnormal state. The change trend is the numerical value and change direction of the trend evaluation metric R(t). The spatial distribution characteristics include the shape characteristics and expansion trend of the hot spot area. The abnormal recognition unit not only considers the static temperature threshold judgment but also incorporates the dynamic trend analysis results. Through the comprehensive evaluation of these characteristics, the accurate recognition and classification of the abnormal state of the equipment are realized, providing a reliable basis for early warning decision-making.

[0165] Generate early warning information according to the abnormal recognition result, including the temperature anomaly level, the location of the abnormal area, the maximum temperature value, and the temperature change trend, and push the early warning information to the system management module.

[0166] For example, the temperature anomaly level is the same as the level corresponding to the acquisition control unit in the image acquisition module, and the specific settings are: normal: T < T1; mild anomaly: T1 ≤ T < T2; moderate anomaly: T2 ≤ T < T3; severe anomaly: T ≥ T3.

[0167] Generate an early warning report according to the abnormal recognition result, including key information such as the temperature anomaly level (normal, mild anomaly, moderate anomaly, severe anomaly), the precise location coordinates of the abnormal area, the current maximum temperature value, and the temperature change trend based on the trend evaluation metric R(t).

[0168] In this embodiment, by combining historical data and real-time temperature data, the temperature change trend of the power station cabinet equipment and the expansion of the hot spot area can be accurately predicted, significantly improving the accuracy of temperature anomaly detection and early warning. Generate the temperature anomaly level based on the temperature field distribution information, historical data, and trend prediction, and generate temperature early warning information in a timely manner to help the operation and maintenance personnel quickly identify the potential failure risks of the equipment and take corresponding measures. It not only improves the early warning ability of equipment failures but also effectively avoids equipment damage and accidents caused by temperature anomalies.

[0169] In an exemplary embodiment, as shown in Figure 3, the present application provides a temperature early warning system for power station cabinet equipment, which includes:

[0170] An image acquisition module 301 for real-time acquisition of dual-light images of the cabinet equipment, that is, infrared thermal imaging images and visible light images;

[0171] An image processing module 302 for preprocessing, feature extraction, spatial registration, and information fusion of the acquired dual-light images to generate a fused image;

[0172] The temperature monitoring module 303 is used to reconstruct the temperature field and detect hot spots on the fused image;

[0173] Intelligent analysis module 304, used to perform trend analysis on the output data of the temperature monitoring module, perform abnormality identification and generate early warning information;

[0174] The system management module 305 is used to implement monitoring data storage, abnormal event recording, warning information push and remote access control, and provide a human-computer interaction interface.

[0175] Among them, the image acquisition module 301 includes: an infrared image acquisition unit, which is used to control the infrared camera parameters and acquire infrared thermal imaging images; a visible light image acquisition unit, which is used to control the visible light camera parameters and acquire visible light images; an acquisition control unit, which is used to receive real-time feedback from the intelligent analysis module 304 through the system management module 305 to realize the acquisition triggering and parameter adjustment of the dual-light camera.

[0176] The image processing module 302 includes: a preprocessing unit, which is used to perform quality assessment, image enhancement and region segmentation on the collected infrared thermal imaging images and visible light images; a feature extraction unit, which is used to extract multi-scale local features and regional features from the preprocessed bi-optical images and construct feature descriptors; an image registration unit, which is used to register the bi-optical images using a hierarchical registration strategy; and an information fusion unit, which is used to fuse the registered bi-optical images to generate a fused image.

[0177] The temperature monitoring module 303 includes: a temperature field reconstruction unit for reconstructing the temperature field based on the fused image; a hot spot detection unit for calculating local statistical features based on the temperature field, using a threshold method to determine the hot spot area, and extracting characteristic parameters of the hot spot area.

[0178] The intelligent analysis module 304 includes: a data processing unit, which is used to time-sequence the temperature field and hot spot detection results output by the temperature monitoring module, establish a monitoring data sequence, spatially weight the data in combination with the regional importance weight, and generate a regional temperature feature vector, including the highest temperature, average temperature, temperature standard deviation and hot spot area; a trend analysis unit, which is used to calculate the temperature change rate and acceleration using a sliding time window method, use exponential smoothing to predict the temperature trend, and establish a trend evaluation index; an anomaly recognition unit, based on multi-level temperature thresholds and trend evaluation indicators, constructs an anomaly discrimination model, and the discrimination basis includes the degree of temperature exceeding the limit, the duration of exceeding the limit, the change trend and the spatial distribution characteristics, and outputs the temperature anomaly level; an early warning generation unit, which generates early warning information according to the anomaly recognition result, including the temperature anomaly level, the location of the abnormal area, the highest temperature value and the temperature change trend, and pushes the early warning information to the system management module 305.

[0179] The system management module 305, as the control center of the entire early warning system, is responsible for realizing the storage management of monitoring data, the recording and tracking of abnormal events, and the timely push of early warning information. At the same time, it provides a human-computer interaction interface. This module receives the real-time feedback from the intelligent analysis module 304 and automatically adjusts the acquisition frequency of the acquisition control unit according to the temperature anomaly level (T < T1 is the normal temperature range, T1 ≤ T < T2 is a mild anomaly, T2 ≤ T < T3 is a moderate anomaly, T ≥ T3 is a severe anomaly), corresponding to the reference frequency f0, 2 times the reference frequency, 4 times the reference frequency, and 8 times the reference frequency respectively, and accordingly adjusts the parameter configuration of the dual-light camera. In addition, this module also provides a remote access control function, enabling operation and maintenance personnel to view the device status, adjust system parameters, and respond to abnormal early warnings at any time, realizing the intelligent management and control of the entire early warning system.

[0180] In this embodiment, (1) through the adaptive acquisition control, multi-feature fusion registration, information fusion with regional importance weighting, and multi-level early warning mechanism based on trend analysis, the accuracy of temperature monitoring of power station cabinet equipment is improved, the timely early warning of temperature anomalies is realized, and a strong guarantee is provided for the safe operation of the power system; (2) an adaptive acquisition frequency adjustment mechanism based on the temperature anomaly level is adopted to dynamically adjust the acquisition frequency according to the device temperature state, optimizing the utilization of system resources while ensuring the monitoring effect, improving the system operation efficiency, and enhancing the system's response speed to abnormal situations; (3) by constructing a feature descriptor containing local features, context information, and prior constraints, combined with a hierarchical registration strategy, the accuracy and robustness of dual-light image registration are improved, effectively solving the problem of difficult image registration in complex environments; (4) an adaptive fusion method based on regional importance weights and temperature gradient information makes full use of the complementary characteristics of dual-light images, improving the quality of the fused images and providing a reliable data basis for subsequent temperature monitoring; (5) a temperature field reconstruction model considering local statistical features, combined with a hot spot detection method with regional importance weights, improves the accuracy of temperature monitoring and can effectively identify and locate temperature anomaly regions; (6) a multi-level early warning mechanism based on time-series data analysis realizes the early warning of temperature anomalies through trend evaluation indicators and multi-dimensional discrimination models, improving the accuracy and timeliness of system early warning and providing an effective decision-making basis for equipment maintenance.

[0181] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0182] Based on the same inventive concept, the embodiment of the present application also provides a power station cabinet equipment temperature warning device for implementing the power station cabinet equipment temperature warning method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more power station cabinet equipment temperature warning device embodiments provided below can refer to the limitations of the power station cabinet equipment temperature warning method above, and will not be repeated here.

[0183] In an exemplary embodiment, as shown in FIG4 , a temperature early warning device for a power station cabinet is provided, comprising: an image acquisition module 401, an image fusion module 402, a temperature detection module 403 and a temperature analysis module 404, wherein:

[0184] The image acquisition module 401 is used to acquire the bi-optical image of the power station cabinet equipment to be analyzed according to the acquisition parameters corresponding to this acquisition; the bi-optical image includes an infrared thermal imaging image and a visible light image;

[0185] An image fusion module 402 is used to perform preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image;

[0186] The temperature detection module 403 is used to perform temperature field reconstruction and hot spot detection on the fused image to obtain temperature field distribution information and hot spot area information;

[0187] The temperature analysis module 404 is used to perform temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information, and obtain temperature warning information of the power station cabinet equipment.

[0188] In one of the embodiments, the temperature warning device for power station cabinet equipment also includes an acquisition parameter determination module, which is used to use the acquisition parameters corresponding to the lowest level of temperature anomaly level as the acquisition parameters for this collection when this collection is the first collection; when this collection is not the first collection, use the acquisition parameters corresponding to the temperature anomaly level determined in the previous collection as the acquisition parameters for this collection.

[0189] In one embodiment, the image fusion module 402 is further used to preprocess the bi-optical image to obtain a preprocessed bi-optical image; the preprocessing includes at least quality assessment processing, image enhancement processing and region segmentation processing; multi-scale local features and regional features are extracted from the preprocessed bi-optical image to construct feature description information; according to the feature description information, the preprocessed bi-optical image is registered to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image; according to the affine transformation matrix, the preprocessed bi-optical image is fused to generate a fused image.

[0190] In one of the embodiments, the image fusion module 402 is further used to select initial registration points from feature points corresponding to feature description information based on the regional segmentation result of the bi-optical image; the regional segmentation result is obtained through regional segmentation processing; candidate registration points are selected from the initial registration points based on the feature description information; the random sampling consistency model is used to eliminate erroneous registration points from the candidate registration points to obtain target registration points; and the affine transformation matrix is ​​calculated and determined based on the target registration points.

[0191] In one of the embodiments, the image fusion module 402 is further used to obtain a registered infrared thermal imaging image based on the affine transformation matrix and the infrared thermal imaging image; obtain an edge map of the visible light image, and obtain a temperature gradient map of the registered infrared thermal imaging image; determine the fusion weight of the dual light image based on the edge map and the temperature gradient map; and fuse the registered infrared thermal imaging image and the visible light image based on the fusion weight to obtain a fused image.

[0192] In one of the embodiments, the temperature detection module 403 is also used to obtain initial temperature field distribution information based on the registered infrared thermal imaging image; obtain temperature field distribution information based on the initial temperature field distribution information, the temperature gradient map, the fused image, the preset reference temperature and the preset power station cabinet equipment area importance weight; determine the hot spot area information based on the temperature field distribution information and the regional segmentation result of the dual-light image by using a threshold judgment method; the regional segmentation result is obtained through regional segmentation processing.

[0193] In one of the embodiments, the temperature analysis module 404 is also used to obtain historical temperature field distribution information and historical hot spot area information corresponding to the historical collection; determine the temperature change trend prediction information and hot spot area expansion trend information based on the historical temperature field distribution information, historical hot spot area information, temperature field distribution information and hot spot area information; determine the temperature anomaly level of this collection based on the temperature change trend prediction information, hot spot area expansion trend information and temperature field distribution information; generate temperature warning information for power station panel cabinet equipment based on the temperature anomaly level, temperature change trend prediction information, hot spot area expansion trend information and temperature field distribution information collected this time.

[0194] Each module in the above-mentioned power station cabinet equipment temperature warning device can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0195] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG5. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a temperature warning method for a power station panel cabinet device is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0196] Those skilled in the art will understand that the structure shown in FIG. 5 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0197] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0198] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0199] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0200] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0201] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0202] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0203] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A temperature early warning method for power station cabinet equipment, characterized in that: The method comprises: According to the acquisition parameters corresponding to this acquisition, a dual-light image of the power station cabinet equipment to be analyzed is acquired; the dual-light image includes an infrared thermal imaging image and a visible light image; Performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image; Performing temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information; According to the temperature field distribution information and the hot spot area information, the temperature of the power station cabinet equipment is analyzed to obtain temperature warning information of the power station cabinet equipment.

2. The method according to claim 1, characterized in that: Before acquiring the bi-optical image of the power station panel cabinet equipment to be analyzed according to the acquisition parameters corresponding to this acquisition, the following is also included: In the case where the current acquisition is the first acquisition, the acquisition parameters corresponding to the lowest level of temperature anomaly are used as the acquisition parameters of the current acquisition; In the case that the current acquisition is not the first acquisition, the acquisition parameters corresponding to the temperature anomaly level determined in the previous acquisition are used as the acquisition parameters of the current acquisition.

3. The method according to claim 1, characterized in that The step of performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image includes: Preprocessing the bi-optical image to obtain a preprocessed bi-optical image; the preprocessing at least includes quality assessment processing, image enhancement processing and region segmentation processing; Extracting multi-scale local features and regional features from the preprocessed bi-optical image to construct feature description information; According to the feature description information, registering the preprocessed bi-optical image to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image; The pre-processed bi-optical image is fused according to the affine transformation matrix to generate a fused image.

4. The method according to claim 3, characterized in that The step of performing registration processing on the preprocessed bi-optical image according to the feature description information to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image includes: According to the region segmentation result of the bi-optical image, initial registration points are screened out from the feature points corresponding to the feature description information; the region segmentation result is obtained by the region segmentation processing; Filtering candidate registration points from the initial registration points according to the feature description information; A random sampling consistency model is used to eliminate erroneous registration points from the candidate registration points to obtain target registration points; The affine transformation matrix is ​​calculated and determined according to the target registration point.

5. The method according to claim 3, characterized in that: The step of fusing the pre-processed bi-optical image according to the affine transformation matrix to generate a fused image comprises: Obtaining a registered infrared thermal imaging image according to the affine transformation matrix and the infrared thermal imaging image; Acquire an edge map of the visible light image, and acquire a temperature gradient map of the registered infrared thermal imaging image; Determining a fusion weight of the bi-optical image according to the edge map and the temperature gradient map; According to the fusion weight, the registered infrared thermal imaging image and the visible light image are fused to obtain a fused image.

6. The method according to claim 5, characterized in that The step of performing temperature field reconstruction processing and hotspot detection processing on the fused image to obtain temperature field distribution information and hotspot area information includes: Obtaining initial temperature field distribution information according to the registered infrared thermal imaging image; The temperature field distribution information is obtained according to the initial temperature field distribution information, the temperature gradient map, the fused image, the preset reference temperature and the preset power station panel cabinet equipment area importance weight; According to the temperature field distribution information and the regional segmentation result of the bi-optical image, the hot spot region information is determined by adopting a threshold judgment method; the regional segmentation result is obtained through the regional segmentation process.

7. The method according to claim 1, characterized in that The step of performing temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature warning information of the power station cabinet equipment includes: Obtain the historical temperature field distribution information and historical hot spot area information corresponding to the historical collection; Determine temperature magnitude change trend prediction information and hot spot area expansion trend information based on the historical temperature field distribution information, the historical hot spot area information, the temperature field distribution information and the hot spot area information; Determine the temperature anomaly level collected this time according to the temperature magnitude change trend prediction information, the hot spot area expansion trend information and the temperature field distribution information; Temperature warning information for the power station panel cabinet equipment is generated based on the temperature anomaly level collected this time, the temperature change trend prediction information, the hot spot area expansion trend information and the temperature field distribution information.

8. A temperature warning device for power station cabinet equipment, characterized in that: The device comprises: An image acquisition module, used to acquire a bi-optical image of the power station cabinet equipment to be analyzed according to acquisition parameters corresponding to this acquisition; the bi-optical image includes an infrared thermal imaging image and a visible light image; An image fusion module, used for performing preprocessing, feature extraction, spatial registration and information fusion on the bi-optical image to generate a fused image; A temperature detection module is used to perform temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot area information; The temperature analysis module is used to perform temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature warning information of the power station cabinet equipment.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images